Deep vein thrombosis (DVT) is a common and high-risk complication among critically ill patients [1]. Its occurrence not only increases the risk of fatal pulmonary embolism, but also prolongs hospital stay, raises medical costs, and adversely affects both prognosis and quality of life [2]. Evidence indicates that critically ill patients are predisposed to DVT due to prolonged immobility, sedation, invasive procedures, and a hypercoagulable state [3]. Despite the widespread usage of traditional preventive strategies, including anticoagulant therapy, mechanical prophylaxis, and routine nursing interventions, their effectiveness is often limited by inadequate patient adherence and resource constraints [4].

The Integrated Theory of Health Behavior Change (ITHBC) emphasizes improving knowledge, enhancing self-efficacy, and reinforcing health beliefs to facilitate sustainable health behaviors [5, 6]. This theoretical framework has been applied in chronic disease management and rehabilitation care, where structured interventions have shown to improve adherence and self-management [7]. While ITHBC has demonstrated efficacy in chronic disease management, its application in acute critical care settings remains limited [7]. Although theory-based interventions are rarely evaluated in intensive care unit (ICU) settings and the high stakes of complication prevention [8], research on their application in critical care, particularly in preventing high-risk complications such as DVT, remains scarce. The rapid advancement of intelligent nursing technologies has enabled the development of digital, individualized care models that leverage predictive analytics and personalized learning algorithms, moving beyond reminder-based interventions [9]. However, no previous studies have attempted to integrate ITHBC’s comprehensive behavior change framework with the intelligent nursing capabilities for DVT prevention. Intelligent interactive platforms can provide real-time risk assessment, personalized intervention pathways, and dynamic adherence monitoring, offering new opportunities for clinical practice [10]. Recent studies have demonstrated the efficacy of theory-driven interventions combined with intelligent technologies in improving patient adherence and clinical outcomes in high-risk settings [11]. Integrating the ITHBC framework with intelligent nursing interventions may enhance awareness, engagement, and compliance of patients and caregivers, thereby strengthening the effectiveness and sustainability of DVT prevention strategies.

Therefore, this retrospective study aims to investigate the clinical application of intelligent nursing interventions guided by the ITHBC for preventing DVT in critically ill patients. By analyzing incidence rates, adherence, and related risk indicators before and after intervention, this study seeks to provide evidence for improving DVT prevention in critical care settings and to inform the broader implementation of intelligent nursing models in intensive care management.

Deep vein thrombosis (DVT) is a frequent and severe complication among critically ill patients, with its development involving the classical triad of hypercoagulability, venous stasis, and endothelial injury [13, 14]. Critically ill patients are particularly susceptible due to prolonged immobilization, mechanical ventilation, sedation, and invasive procedures, all of which contribute to impaired venous return and coagulation abnormalities [15]. Once thrombi form and embolize, life-threatening pulmonary embolism may occur. Current preventive strategies mainly include pharmacological prophylaxis, such as low-molecular-weight heparin, and mechanical methods such as intermittent pneumatic compression and elastic stockings [16]. However, pharmacologic prophylaxis carries bleeding risks, and efficacy of mechanical approaches relies heavily on patient adherence, which often limits their effectiveness in practice [17, 18]. Thus, optimizing the safety, efficacy, and sustainability of DVT prevention remains a critical challenge in critical care nursing.

The present study demonstrated that the incidence of DVT was significantly lower in the observation group than in the control group, along with notable improvements in nursing adherence and patient satisfaction, without a significant increase in complication rates. These findings suggest that intelligent nursing interventions guided by the ITHBC provide a valuable supplement to traditional prevention strategies in critically ill patients. Improved adherence played a central role in reducing the risk of DVT. Through real-time risk assessment and personalized reminders, the intelligent nursing platform enhanced patients’ and caregivers’ understanding of DVT risk factors and preventive strategies. Multimedia education combined with mobile-based alerts overcame the limitations of traditional one-way health education, empowering patients to actively participate in DVT prevention efforts. Previous studies have emphasized that poor adherence is a major barrier to effective DVT prevention [19, 20]; our results confirm that a theory-based, technology-assisted nursing model can effectively address this gap. Regarding hematological indicators, the observation group showed significantly lower D-dimer levels after intervention compared with the control group, while PT and APTT levels remained comparable. D-dimer, a fibrin degradation product, is a sensitive biomarker of thrombus formation and lysis [21]. Its reduction reflects a decreased risk of thrombosis [22, 23]. This suggests that intelligent nursing interventions, by promoting frequent repositioning, encouraging limb exercises, and increasing adherence to mechanical prophylaxis, may improve venous hemodynamics and indirectly mitigate hypercoagulability.

In terms of safety, the incidence of complications such as bleeding and skin pressure injuries was slightly lower in the observation group, with no statistically significant difference. This indicates that intelligent nursing interventions are both safe and feasible in critically ill patients. Moreover, patient satisfaction scores were significantly higher in the observation group, reflecting that individualized and technology-assisted care not only improves clinical outcomes but also enhances the care experience, thereby fostering stronger patient–caregiver relationships.

This study demonstrates distinct innovations compared with conventional DVT prevention research. While prior studies have predominantly focused on pharmacological efficacy and mechanical device optimization [24], our approach integrates behavioral theory with intelligent technology in a novel framework. Our ITHBC-guided approach achieved a more substantial reduction (7.7% vs. 18.1%, p 0.05) by addressing the psychological determinants of adherence, and our intelligent platform provided continuous, adaptive prevention that responded to real-time patient data. This represents a paradigm shift from static protocols to dynamic, patient-tailored prevention.

By incorporating ITHBC into critical care practice, this study highlights the importance of reinforcing the knowledge-belief-behavior pathway while leveraging intelligent platforms to achieve closed-loop management. The present study observed significant improvement in patient satisfaction (96.5% vs. 87.7%, p 0.05). This indicates that the ITHBC’s emphasis on partnership and selfefficacy may enhance technology acceptance. Consistent with existing literature, our findings further validate the effectiveness [25].

This study, however, has certain limitations. It was a single-center retrospective study with a relatively small sample size, which may introduce selection bias. The follow-up period was relatively short, limiting the evaluation of long-term outcomes. Furthermore, patient-level heterogeneity, such as variations in nutritional status, inflammatory markers, and genetic predispositions, was not comprehensively analyzed. Therefore, future multicenter, large-sample, prospective studies with extended follow-up are needed to further validate these findings and refine individualized nursing strategies.

Intelligent nursing interventions guided by the ITHBC effectively reduced the incidence of DVT, improved adherence, and enhanced patient satisfaction in critically ill patients, without raising the incidence of complications. This model offers a novel, safe, and practical approach for DVT prevention in critical care, and holds promise for broader clinical application.

• Intelligent nursing interventions guided by the Integrated Theory of Health Behavior Change (ITHBC) significantly reduced deep vein thrombosis (DVT) incidence in critically ill patients.

• The intervention model improved nursing adherence through real-time monitoring and personalized reminders.

• Patient satisfaction was enhanced with ITHBC-guided intelligent nursing interventions, while safety was maintained, with no significant increase in complications.

• This approach represents an innovative integration of behavioral theory with intelligent healthcare technology.

• The study provides evidence for implementing theory-driven nursing models in critical care settings.

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. The intelligent nursing platform software is proprietary and available for research collaboration through institutional agreements.

CWL and PX: conceptualization, methodology, writing-original draft. JH and YQZ: data curation, formal analysis. MYL, YQZ and HW: investigation, resources. QL and HW: software, validation. All authors contributed to revising the manuscript critically for important intellectual content. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was approved by the Medical Ethics Committee of The Affiliated Xuzhou Municipal Hospital of Xuzhou Medical University (Approval No.: xyyll[2025]173). Informed consent was waived due to the retrospective nature of the study and the use of anonymized data. The study adhered to the Declaration of Helsinki and relevant institutional guidelines. All procedures were conducted in accordance with relevant guidelines and regulations, with strict measures taken to ensure patient confidentiality throughout the study.

The authors thank the nursing staff of the ICU for their assistance in data collection and implementation of the interventions.

This research received no external funding.

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/BJHM52246.